A parametric method to estimate environmental energy efficiency with non-radial adjustment: an application to China

To estimate the performance of China in terms of energy use efficiency during the first two decades of the twenty-first century while also taking into consideration pollutant emission, this study uses a panel data set covering 30 provincial administrative regions in mainland China for the period 2000–2016. To overcome problems with the DEA-based method, this study proposes an SFA-based model that can estimate environmental energy efficiency while maintaining the regularity constraints imposed on undesirable output, by using Bayesian technique. Our empirical results show that the average value of environmental energy efficiency during the whole sample period changed from 0.7858 in 2000 to 0.7726 in 2016, with an average value of 0.7812 over the whole period. This result is in sharp contrast with findings based on the often-used GDP/energy and GDP/undesirable output indexes, both of which show an improving trend over same sample period. This study suggests that more sophisticated indexes should be used to evaluate meaningful energy efficiency and environmental protection-related performance.